This post was written by Clayton Crawford, Product Engineer on the 3D Analyst team in ESRI's Software Products group in Redlands

Due to the demand for lidar training, ESRI’s Training Center now offers three web-based courses on lidar. First is a free training seminar that provides an overview of lidar capabilities in ArcGIS and introduces high level concepts. The other two include hands-on exercises and are geared toward data managers and analysts. For more information see each course’s description...

Getting Started with Lidar in ArcGIS 

Managing Lidar Data in ArcGIS

Using Lidar Data in ArcGIS

This is the seventh blog in a series on Lidar Solutions in ArcGIS. It was written by Clayton Crawford, Product Engineer on the 3D Analyst team in ESRI’s Software Products group in Redlands.

Lidar is promoted as an accurate form of elevation data. To a large degree, this comes from its dense sample interval. Some refer to lidar as ‘painting the ground with measurements’. Indeed, it’s commonplace to get sub-meter sampling. This facilitates canopy penetration which improves the accuracy of ground models in forested areas. The high sample density also improves results for certain applications such as floodplain delineation. That said, lidar is not a silver bullet for all surface modeling activities. Two areas that tend to be problematic are contour derivation and slope analysis.

Messy contours and steep slopes

Getting a computer to produce nice looking contours from traditional data sources is hard enough. It's even harder to produce them from lidar. Contours produced from full resolution lidar tend to be jagged with many twists and turns and isolated closed rings (Figure 1).

Slope assessment with lidar is also problematic. If one examines the average slope from full resolution lidar they will find it’s unusually high. This is true even over relatively flat ground as shown in Figure 2.

Some people misinterpret these issues with contours and slope as simply a result of lidar being more accurate than other forms of surface data. At large enough scales almost any surface is rough, right? While there’s some truth to this a significant part of the problem really stems from the relationship between horizontal sample density and vertical accuracy. As with any measurement technology lidar is not perfectly accurate. Its vertical accuracy generally ranges from 12 to 15cm. That provides a random height difference between two adjacent points of 24 to 30cm. When the points are only one meter or less apart horizontally that height differential becomes significant. From a signal processing perspective this is called high frequency noise.

Techniques to reduce noise

In order to get less jagged contours and more reasonable slope estimates you have to remove the noise from the lidar and lose as little real information as possible while doing so. While there’s no way to prevent any loss, harm can be minimized. The terrain dataset offers a couple of tools to facilitate this. One is intelligent point thinning; the other is a high quality interpolator.

Point thinning occurs during the terrain pyramid building process. Some people think pyramids are strictly a visualization tool, used solely to speed up drawing. While this is true for raster pyramids, the same does not apply to terrains. Terrains were designed knowing that lidar is better suited for some applications when it’s been generalized.

There are two terrain point thinning algorithms: z-tolerance and windowsize. Both have something to offer and are arguably better than random n^th point filters used by other solutions that, while fast and reasonable for visualization, are not what you’d want to use for analysis.

The z-tolerance filter employs a TIN based algorithm to find a subset of points sufficient to create a surface that’s within a given vertical distance to the full resolution surface. TerraScan, an industry leading lidar processing package, uses a similar technique for the selection of what it refers to as ‘model key’ points. The use of this filter is recommended when a quantifiable measure of vertical accuracy is needed.

The windowsize filter selects points within a given horizontal sample distance. Every so many units in XY (i.e., the sample window size) the points within that area are examined and one or two are selected depending on which option you chose: the one closest to the mean of the other points in the sample window, the highest or lowest of those points, or both the highest and lowest. As stated earlier, much of the noise problem results from a bad ratio of high horizontal sample density to vertical accuracy. The windowsize filter enables you to improve that ratio. While it’s hard to quantify the accuracy of the thinned data, empirical evidence has shown this approach works well.

Thinning the data

Lidar points are thinned when a terrain dataset is built. The filter algorithm applied is based on the type of pyramid selected for the terrain. Fortunately, the same name is used for pyramid type as filter algorithm so there’s no ambiguity. The pyramid type is selected via the terrain wizard found on the feature dataset context menu in ArcCatalog (Figure 3).

The next step is to specify the pyramid levels. For the sake of noise reduction we’re interested in that first pyramid level which is one step removed from full resolution (Figure 4). A z-tolerance equal to the vertical accuracy of the data is reasonable for that level. This will eliminate as much noise as possible while decreasing the accuracy of the result as little as possible. If you start out with 15cm vertical accuracy, you end up with approximately 30cm accuracy and if you go on to produce contours, set their interval to at least double this, namely 60cm or approximately 2 feet. When building a windowsize pyramid a sample distance equal to twice the nominal point spacing is reasonable.

After having built a terrain, the thinned points reside in its pyramid. You can extract a TIN made from these thinned points via the Terrain to TIN tool and from this generate contours and slope. The results will be less noisy than if made from the un-thinned points. One drawback to this approach is that TINs have a size restriction. Ten million points per TIN is about as large as you can go. With lidar projects this can mean that you have to extract multiple TINs from a single terrain, even when using thinned data, if you’re working on a large area. Another drawback is that contours and slope estimates made from these TINs will still be more angular and discontinuous than necessary. A recommended alternative is to go through a raster.

Note, in ArcGIS 9.4 you will be able to derive contour and slope output directly from terrains without needing to go through TINs.

Raster Interpolation

Contour and slope output can be improved by deriving them from a raster that’s made from a terrain. Engineers have a traditional bias towards working directly on TINs, so they may not like the idea of going through a raster, but when dealing specifically with contour and slope production from lidar this bias is unwarranted. For one thing, lidar points are essentially collected in a random distribution and the resulting triangles are not handpicked or guaranteed to fit some mathematical model made inside a CAD package (e.g., for road design). Additionally, the linear piecewise surface defined by planar triangle faces is not smooth. A smoother surface can be made using the Terrain To Raster tool with the natural neighbors interpolation option. An added benefit of using Terrain To Raster is that you can rasterize an entire terrain in one shot and avoid the size constraint associated with TIN extraction (Figure 5).

You then use the Contour and Slope tools on the derived raster. Figures 6 and 7 illustrate the difference between full resolution and generalized contour and slope derivatives.

Conclusion

Lidar is noisy because of a low vertical accuracy relative to horizontal sample distance. This noise translates into poor quality contours and excessively high average slope rates. Noise can be reduced via point thinning and smoothing without also losing much of the accuracy and detail. The terrain dataset provides these tools via pyramiding and natural neighbor interpolation.

That’s it for this installment of the Lidar Solutions in ArcGIS series. Subscribe to this blog or check back in a month or so for a look at business partner offerings.

The 2009 User Conference has just ended.  We had a great time and it was wonderful to meet with everyone and talk about geoprocessing and analysis.  We had some great sessions and really interesting questions at the geoprocessing and analysis island. 

We'll be bundling up our tech session slides, models, data, and demo scripts and submitting them to the Model and Script tool gallery.  You should start seeing submissions in a few days -- they'll be tagged with '2009 User Conference'.  In addition, some session videos will become available in a few weeks as our vendor processes them.  Check the Media Gallery for these session videos.

This is the sixth blog in a series on Lidar Solutions in ArcGIS. The author of this blog is Clayton Crawford, lead Product Engineer on ESRI's 3D Analyst team in the Software Products group in Redlands.

Updating a portion of a terrain dataset with new measurements

The ability to update a surface is important to people responsible for providing accurate, up to date surfaces and people performing analysis on those surfaces. Updates come in different forms:

• Adding ancillary data (e.g., breaklines)
• Removing or replacing bad data
• Using newer or more accurate data
• Increasing extent with additional data
• Inserting design/modeled data to perform ‘what-if’ analysis

These kinds of updates are best performed on the measurements used to construct a surface rather than on derivatives like raster DEMs. Those can be recreated as needed after the measurement edits have taken place. Terrain datasets support this editing model because they maintain a direct link to the source measurement data. When you modify the measurements, you are automatically modifying the terrain in the same process.

How terrains are edited

Editing a terrain dataset is really about editing measurements. Using standard feature edit tools, you can manipulate the measurements that reside in feature classes that participate in a terrain.

Terrains are made from one or more feature classes with some simple rules for each that control how they get used to shape the terrain surface. For example, a multipoint feature class containing lidar points can get added as mass points, a line feature class containing streams and lake shores is used as a source of breaklines, and a polygon feature class can control the data area boundary.

Most feature classes used to define a terrain are what we call referenced. This means that the terrain maintains a pointer, or handle, to them. The terrain prevents its referenced feature classes from being deleted, and pays attention to any edits that occur on them including the addition, deletion, or modification of feature geometry. You can use the feature editor in ArcMap as well as geoprocessing tools to modify these feature classes. A terrain will automatically flag itself as ‘dirty’ in areas where edits were made. Then the terrain can be rebuilt to bring its pyramid in sync with the updated features. It does this based on the dirty areas so it’s a local, or partial, process; the entire terrain does not need to be reconstructed.

Multipoint feature classes have the option of being embedded. When multipoints are embedded the terrain build process copies the points into pyramid tables held private by the terrain and it becomes the container for the points. The terrain does not reference the source feature class. That can be deleted, allowing you to retrieve what is typically a substantial amount of disk space; approximately 1GB per 150 million points. Terrain specific tools, Add Terrain Points (which can both add and replace) and Remove Terrain Points, are used to edit the embedded points based on an area of interest. These tools also offer the benefit of being BLOB attribute aware so if you have any LAS attributes stored with those multipoints (for more on this topic see blog on creating intensity images) the tools know how to keep the BLOB based values in sync with the points relative to the edits. For example, if a few vertices of a multipoint are deleted from an embedded feature class, the terrain will delete the corresponding BLOB based attribute values for those points.

Appending measurements

Measurements can be added to a terrain via the Append and Add Terrain Points geoprocessing tools. The Append tool operates on regular (referenced) feature classes. The Add Terrain Points tool is used to add or replace points in embedded feature classes. You can also add a feature class to an existing terrain via the Add Feature Class To Terrain tool but be aware that this is treated as a schema edit that invalidates the entire terrain, requiring a full rebuild. If data is to be added incrementally, it’s best to append it to a feature class that already participates in the terrain than add a new feature class to the terrain for each new set of data.

Let’s take a scenario where data is provided in phases. In this example the bare earth lidar points are made available first. The breaklines come later in several deliveries. Knowing this schedule, you can create a terrain referencing the lidar multipoint feature class plus an empty line feature class held in anticipation of the breaklines. See Figure 1 with a zoomed-in view of a terrain made solely with bare earth lidar points.

When some of the breaklines are made available they are added to the terrain by adding them to the line feature class referenced by the terrain. This is done using the geoprocessing Append tool (Figure 2).

After running Append, the terrain will become ‘dirty’ in the areas where the lines were added. To see the dirty areas you add a dirty area renderer from the Symbology tab of the terrain layer Properties dialog (Figures 3a and 3b).

At this point the terrain needs to be rebuilt. This is done using either the Build Terrain geoprocessing tool or the Build button on the Update tab of the terrain Properties dialog in ArcCatalog. Once the terrain is re-built the improvement made by the breaklines is evident (Figure 4).

Replace

With lines and polygons in referenced feature classes the replacement of measurements is a two step process. First you delete the old then append the new. If you’re only dealing with a handful of features then you can select and delete them using the Editor in ArcMap. For larger collections rely on geoprocessing tools. For example, use Select By Location followed by Delete Features and Append.

It’s easiest to replace lidar points if they’re embedded. There’s a geoprocessing tool called Add Terrain Points that has a Replace option. This will replace all the points within a given area. So, if you discover something was wrong with a few source point files that were used to build a terrain you can replace them without needing to rebuild the entire terrain from scratch. Figure 5 shows an example where one area, in an otherwise bare earth model, that was inadvertently loaded with first return data.

To fix this problem load the replacement data into a new multipoint feature class. Then run the Add Terrain Points tool with the Replace option. By default, the replacement area will come from the extent of the input feature class (Figure 6).

Once the points have been replaced the terrain needs to be rebuilt to update the affected area. Run the Build Terrain geoprocessing tool or use the Build Terrain button on the Update tab of the terrain Properties dialog in ArcCatalog – either one fixes the terrain.

Conclusion

There are times when updates to a surface model are needed, be it for quality improvement or what-if scenario analysis. It’s hard to make these types of updates to derivative products like raster DEMs without ending up with some anomalies around the update area. It’s more appropriate to modify the source measurement data from which the surface model is derived. For larger datasets, like those coming from lidar, it’s also desirable that datasets only be reprocessed where the updates occur rather than rebuilding everything. Terrain datasets support this by maintaining links to their source measurements in the geodatabase and their use of dirty areas.

That concludes part 6 of Lidar Solutions in ArcGIS. Subscribe to this blog or check back in a month or so for a discussion on how to minimize noise from lidar for contouring and slope analysis.

This blog post was written by David Wynne, Product Engineer on the Geoprocessing Team in ESRI’s Software Products group in Redlands.

The Python Community announced the release of Python 2.5.4 late in December of 2008. Python 2.5.4 is a bug fixes release.

While ArcGIS 9.3.1 was shipped with Python 2.5.1, Python 2.5.4 has now been certified and is supported. In limited testing with ArcGIS 9.3, no issues were encountered.  To take advantage of fixes in Python 2.5.4, it can simply be installed on top of Python 2.5.1. That means, it is not necessary to uninstall Python 2.5.1 first.* 

To see and to download the release notes go here.

* If you uninstalled Python 2.5.1 and you work with the Spatial Statistics geoprocessing tools, you must reinstall Numerical Python (NumPy) from the ArcGIS installation DVDs.

For more information, see FAQ: Can Python 2.5.4 be used with ArcGIS?

This blog post was written by Lauren Scott, Geoprocessing/Spatial Statistics Product Engineer in the Software Products Group at ESRI in Redlands.

There are a number of good resources where you can learn about the new regression analysis tools in ArcGIS 9.3.  The newest resource is a free one-hour Web seminar,  Regression Analysis Basics in ArcGIS 9.3, available for download from the ESRI Virtual Campus.

 Other Regression Analysis resources include:

• Regression Analysis Basics Documentation
• Ordinary Least Squares Regression Tool Help and Learn more about how OLS works
• Geographically Weighted Regression Tool Help and Learn more about how GWR works
• Interpreting OLS tool results
• Interpreting GWR tool results
• Several articles can be found in the Spring 2009 issue of ArcUser magazine [Try the Flip Page view of ArcUser!]:
• Answering Why Questions: An introduction to using regression analysis with spatial data;
• Regression Analysis Components;     and
• Applying Geographically Weighted Regression to a Real Estate Problem.
• GEOconnexion International Magazine, December/January 2009, includes an article titled Regression Analysis Tools for GIS Modeling
• Chapter 5, Analyzing Geographic Relationships, in The ESRI Guide to GIS Analysis, Volume 2 discusses regression analysis and some of the challenges when it is applied to spatial data.

Enjoy!

You may not have noticed, but you can send us feedback about this site. There's a feedback link at the bottom right of the home page, as illustrated at the bottom of this post.

We're particularly interested in hearing from you about blog post topics you'd like to see and suggestions on how we could make this site more useful.

With the Calculate Field tool, you can easily create expressions that use some property of a feature's shape, such as length or area. You can also convert length and area to different units. The illustration below shows calculating the MILES field to the length (in miles) of each line feature.

Details about the Expression syntax

The expression syntax -- !shape.length@miles! -- starts and ends with the special 'decode field' character. This decode field character is different depending on the value of the Expression Type parameter. For Expression Type of PYTHON or PYTHON_9.3, the character is an exclamation point ("!"), as shown in the illustration above. For VB, the decode field characters are "[" and "]".

Inside the decode field characters, you can put the name of any field found on the input table. You can also use the reserved field name "shape" followed by a period and an attribute of the shape. In this case, we're using the length attribute: !shape.length!

Finally, we convert shape.length to miles with the use of the special unit conversion syntax '@<unit>'. Unit can be any of the following: CENTIMETERS | DECIMALDEGREES | DECIMETERS | FEET | INCHES | KILOMETERS | METERS | MILES | MILLIMETERS | NAUTICALMILES | POINTS | YARDS.

If your shape type is polygon, you can use these areal unit keywords: ACRES | ARES | HECTARES | SQUARECENTIMETERS | SQUAREDECIMETERS | SQUAREINCHES | SQUAREFEET | SQUAREKILOMETERS | SQUAREMETERS | SQUAREMILES | SQUAREMILLIMETERS | SQUAREYARDS

More things you can do with the shape object and its attributes

The shape field is actually a geometry object with the following properties.

If, for example, you wanted the X-coordinate of the first point of a line, you would use !shape.firstpoint.x! To find the minimum X-coordinate of the line, use !shape.extent.xmin!

Simple equations

The expression can contain simple equations. For example, to calculate minutes it would take to traverse a street segment going 30 miles per hour, use:

!shape.length@miles! / (30.0 / 60.0)

 Note that:

!shape.length@miles! / (30 / 60)

won't work - when doing real arithmetic you need the decimal point after 30 and 60.

Code blocks

See http://www.esri.com/news/arcuser/0507/files/pythonscript.pdf for information on using code blocks in Calculate Field